Analytical methods based on high resolution liquid chromatography separation followed by mass spectrometry detection (LC-MS) or UV absorbance (LC-UV) detection are widely used for the analysis of a wide range of compounds such as biotoxins, drugs, persistent environmental pollutants, and other chemicals in wide range of samples such as plant and animal tissues, soil, water, etc. The identification of compounds present in samples is usually based on a match of both chromatographic retention time and mass or UV spectral data for authentic chemical standards with those of putative compounds observed in a sample. However, the absolute retention times of analytes can be highly variable between different laboratories and instruments, and even between days in the same laboratory. This usually requires the analysis of chemical reference standards on the same day each batch of samples is run to allow a good match of retention times for the conclusive identification of potential contaminants. This approach increases the workload of analysts and the cost of analyses. In addition, not every laboratory can stock all standards in order to accomplish this task as analysts may be concerned with the analysis of hundreds of possible analytes. It would be helpful to have a better way of cataloging retention data so that analytes can be more easily identified through a match of retention times without the use of in-house standards.
As an example, the above is particularly true with the analysis of marine and freshwater biotoxins. There are many different groups of biotoxins and within each group there can be many different structural analogues. Standards for many of these biotoxins are not commercially available and if they are, they can be very expensive. Analysts in this field face a difficult problem of determining which biotoxin analogues might be present in samples such as water, algae and shellfish. Similarly, for structural analogues or metabolites of pharmaceuticals and environmental pollutants, very few laboratories have ready access to standard compounds. This would be of concern to fields such as athlete doping control and monitoring of environmental samples and foodstuffs, among others.
Another issue of concern is related to the establishment of routine LC-MS analysis methods such as scheduled selected reaction monitoring, in which specific ion transitions are monitored over narrow windows that encompass the analytes of interest. The first step in setting up such a method is usually to perform an analysis of a mixture of all target analytes prior to establishing the windows. Again, the possible lack of every standard compound in a laboratory, as well as the extensive work required in this operation, presents problems to the analyst.
One way to correct for variations in retention data is to use “relative retention times (RRT)”, in which analytes' retention times are measured relative to that of an internal standard compound. This method works fairly well in isocratic LC (constant solvent composition) but not in the more commonly-used gradient mode (changing solvent composition) because different LC instruments have different hold-up volumes in the gradient mixing system, which results in offsets in RRT values. The RRT values will also vary if there is any difference in the rate of change of the gradient slope or in column dimensions.
A better way to report retention data is to use “retention index (RI)” values. In this procedure, a series of homologous reference compounds are co-injected with the analytes. An interpolation of analyte retention times into a fitted curve of the plot of retention time vs. retention index value for the reference compounds results in a retention index value for each analyte.
The use of retention indices has been widely used in the field of gas chromatography. In this case, a series of n-alkanes is usually used as the RI standards and the resulting interpolated indices are usually referred to as “Kovats retention indices”. This is possible because the commonly used flame ionization detector responds well to most organic compounds, including the n-alkanes, albeit that all the compounds must be volatile. These are not applicable to LC, especially LC-UV or LC-MS, because the n-alkanes are not easily detected by common UV or MS detectors.
Several different RI systems have been investigated by other researchers for use in LC-UV analysis (see Scheme 1).
These are most commonly used with the UV absorbance detector. These include Smith's work on alkyl aryl ketones (Smith R. M., (1982) J. Chromatogr. 236, 313-320; Smith R. M., (1995) Journal of Chromatography Library 57, 93-144), Baker and Ma's work on 2-ketoalkanes (Baker J. K., Ma C-Y., (1979) J. Chromatogr. 169, 107-115) and Bogusz and Aderjan's work on 1-nitroalkanes (Bogusz M., Aderjan R., (1988) J. Chromatogr. 435, 43-53). Two journal papers by one group alluded to the use of parabens (n-alkyl esters of 4-hydroxy benzoic acids) for measuring retention indices of phenols using LC-UV analysis (Yamauchi S., Mori, H., (1990) J. Chromatogr. A. 515, 305-311; Yamauchi S., (1993) J. Chromatogr. 635, 61-70). There have also been publications on application of LC-UV retention indices in the toxins field (Kuronen P., (1989) Archives Environ. Contam. Toxicology 18, 336-48; Frisvad J., Thrane U., (1987) J. Chromatogr. 404, 195-214; Hill D. W., Kelley T. R., Langner K. J., Miller K. W., (1984) Analytical Chemistry 56, 2576-2579). There has been only one publication on the use of retention indices for LC-MS (Kostiainen R., Kuronen P., (1991) J. Chromatogr. 543, 39-47) and this was based on 1-[p-(2,3-dihydroxypropoxyl)phenyl]-1-alkanones (Scheme 1) as retention index standards. However, the Kostiainen method uses a set of standards that are complicated to synthesize and not commercially available, but more importantly are not well suited to modern LC-MS methods based on electrospray (ESI) or atmospheric pressure chemical ionization (APCI).
Retention index standards of the prior art such as the alkyiphenones were designed mainly for LC with UV absorbance detection. They are not well suited to LC-MS methods based on electrospray (ESI) or atmospheric pressure chemical ionization (APCI), the most commonly used ionization systems. In-studies carried out by the Inventors of the present invention, tests on alkylphenones as retention index standards in LC-MS show that the sensitivity of the alkylphenones in positive ion ESI is low, which requires more concentrated solutions to be injected which in turn results in bad peak shapes in the chromatograms. Also, they cannot be detected in the negative ion mode. Tests on parabens as retention index standards in LC-MS were more promising because they can be detected in negative ion mode but overall, their performance was lacking in terms of sensitivity and they still require high concentrations to be injected. In addition, there is potential susceptibility of the phenolic compounds to retention time changes due to variations in pH of the mobile phase, which can change the charge state of the parabens.
There remains a need for effective LC retention index standards, particularly for LC-MS methods.